with open(prms_file_name, 'r') as p_fp: params = ast.literal_eval(p_fp.read()) layers = params['layers'] tr_prms = params['training_params'] ############################################## Load print('Loading Data') try: img_sz = layers[0][1]["img_sz"] except KeyError: img_sz = layers[0][1]["img_sz"] = 32 pad_width = (img_sz-32)//2 data_x, data_y = utils.load_pad_info(data_file, pad_width, 0) corpus_sz = data_x.shape[0] # Print top three layers for lyr, prms in layers[:3]: print(lyr) for param, value in prms.items(): print(" {:20}:{}".format(param, value)) print() ############################################## Init Layer imgs = shared(np.asarray(data_x, config.floatX), borrow=True) x_sym = tt.tensor4('x') net_layers, i = [], 0 lyr_look_up = {"InputLayer":InputLayer, "ElasticLayer":ElasticLayer,
with open(nnet_prms_file_name, 'rb') as nnet_prms_file: params = pickle.load(nnet_prms_file) layers = params['layers'] tr_prms = params['training_params'] allwts = params['allwts'] ############################################# Load Data print("\nLoading the data ...") try: img_sz = layers[0][1]["img_sz"] except KeyError: img_sz = layers[0][1]["img_sz"] = 32 pad_width = (img_sz - 32) // 2 trin_x, trin_y = utils.load_pad_info("train", pad_width, 0) test_x, test_y = utils.load_pad_info("test", pad_width, 0) ############################################# Init Network params['training_params']['BATCH_SZ'] = batch_sz = 100 ntwk = NeuralNet(**params) ############################################# Read glyphs & classify coarse = utils.fine_to_coarse def test_wrapper(tester, truth): print("Classifying...") fine_errors = np.zeros((100, 100), dtype="uint") coarse_errors = np.zeros((20, 20), dtype="uint") sym_err, bit_err, n = 0., 0., len(truth) // batch_sz
with open(prms_file_name, 'r') as p_fp: params = ast.literal_eval(p_fp.read()) layers = params['layers'] tr_prms = params['training_params'] ############################################## Load print('Loading Data') try: img_sz = layers[0][1]["img_sz"] except KeyError: img_sz = layers[0][1]["img_sz"] = 32 pad_width = (img_sz - 32) // 2 data_x, data_y = utils.load_pad_info(data_file, pad_width, 0) corpus_sz = data_x.shape[0] # Print top three layers for lyr, prms in layers[:3]: print(lyr) for param, value in prms.items(): print(" {:20}:{}".format(param, value)) print() ############################################## Init Layer imgs = shared(np.asarray(data_x, config.floatX), borrow=True) x_sym = tt.tensor4('x') net_layers, i = [], 0 lyr_look_up = { "InputLayer": InputLayer,
print('Host :', socket.gethostname()) print(nn.get_layers_info(layers)) print(nn.get_training_params_info(tr_prms)) ########################################## Load Data print("\nLoading the data ...") try: img_sz = layers[0][1]["img_sz"] except KeyError: img_sz = layers[0][1]["img_sz"] = 32 pad_width = (img_sz-32)//2 trin_x, trin_y = utils.load_pad_info("train", pad_width, 0) test_x, test_y = utils.load_pad_info("test", pad_width, 0) batch_sz = tr_prms['BATCH_SZ'] n_train = len(trin_y) n_test = len(test_y) trin_x = share(trin_x) test_x = share(test_x) trin_y = share(trin_y, 'int32') test_y = share(test_y, 'int32') ################################################ print("\nInitializing the net ... ") net = nn.NeuralNet(layers, tr_prms, allwts) print(net)
params = pickle.load(nnet_prms_file) layers = params['layers'] tr_prms = params['training_params'] allwts = params['allwts'] ############################################# Load Data print("\nLoading the data ...") try: img_sz = layers[0][1]["img_sz"] except KeyError: img_sz = layers[0][1]["img_sz"] = 32 pad_width = (img_sz-32)//2 #trin_x, trin_y = utils.load_pad_info("train", pad_width, 0) test_x, test_y = utils.load_pad_info("test", pad_width, 0) test_col = (np.rollaxis(test_x, 1, 4)*255).astype("uint8") ############################################# Init Network params['training_params']['BATCH_SZ'] = 1 ntwk = NeuralNet(**params) tester = ntwk.get_data_test_model(go_nuts=True) ############################################# Image saver dir_name = os.path.basename(nnet_prms_file_name)[:-4] + '/' if not os.path.exists(dir_name): os.makedirs(dir_name) namer = (dir_name + '{}_{:02d}.png').format # Usage:namer(info, i)